Skip to content

Instantly share code, notes, and snippets.

@qh-huang

qh-huang/delete_non-code_files.py

Last active July 24, 2023 11:39
Show Gist options
  • Save qh-huang/f20298fbf1bdf97e4478748bec330bf3 to your computer and use it in GitHub Desktop.
Save qh-huang/f20298fbf1bdf97e4478748bec330bf3 to your computer and use it in GitHub Desktop.
Download ZIP
Trim tools for large code base trim to avoid massive processing time for other software to process
Raw
delete_non-code_files.py
#!/usr/bin/env python3
import os
import sys
from tqdm import tqdm # Import tqdm for the progress bar
def is_code_file(file_path):
code_file_names = {"Makefile", "configure", "Kconfig", ".gitignore", ".gitmodules", "CMakeFile"} # Add more conventional file names as needed
code_file_extensions = {".c", ".cc", ".cpp", ".h", ".hh", ".hpp", ".inc", ".java", ".dts", ".dtsi", ".log", ".conf",
".ini", ".xml", ".yaml", ".xls", ".csv", ".js", ".py", ".s", ".gitmodules", ".cmake", ".gn", ".md", ".m4", ".m", ".diff",
".mk", ".mak", ".sh", ".json", ".txt", ".po"} # Add more extensions as needed
if os.path.basename(file_path) in code_file_names:
return True
_, file_extension = os.path.splitext(file_path)
file_extension = file_extension.lower()
return file_extension in code_file_extensions
def delete_non_code_files(directory_path):
total_files = 0
for root, dirs, files in os.walk(directory_path):
total_files += len(files)
with tqdm(total=total_files, desc="Deleting Non-code Files", unit="file") as pbar:
for root, dirs, files in os.walk(directory_path):
for filename in files:
file_path = os.path.join(root, filename)
if not is_code_file(file_path):
try:
os.remove(file_path)
print(f"Deleted file: {file_path}")
except OSError as e:
print(f"Error deleting {file_path}: {e}")
pbar.update(1)
def display_usage():
print("Usage: python script.py <directory_path>")
print("This script walks through the specified directory and deletes non-code files.")
if __name__ == "__main__":
if len(sys.argv) != 2 or sys.argv[1] in ("-h", "--help"):
display_usage()
sys.exit(1)
directory_path = sys.argv[1]
delete_non_code_files(directory_path)
Raw
remove_empty_directories.py
#!/usr/bin/env python3
import os
import sys
from tqdm import tqdm # Import tqdm for the progress bar
def delete_empty_directories(directory_path):
total_dirs = 0
for root, dirs, files in os.walk(directory_path, topdown=False):
total_dirs += len(dirs)
with tqdm(total=total_dirs, desc="Deleting Empty Directories", unit="dir") as pbar:
for root, dirs, files in os.walk(directory_path, topdown=False):
for dirname in dirs:
dir_path = os.path.join(root, dirname)
if os.path.islink(dir_path):
# If the directory is a symbolic link, just unlink it
os.unlink(dir_path)
print(f"Unlinked symbolic link: {dir_path}")
elif not os.listdir(dir_path): # Check if the directory is empty
try:
os.rmdir(dir_path)
print(f"Deleted empty directory: {dir_path}")
except OSError as e:
print(f"Error deleting {dir_path}: {e}")
pbar.update(1)
def display_usage():
print("Usage: python script.py <directory_path>")
print("This script walks through the specified directory and empty directories.")
if __name__ == "__main__":
if len(sys.argv) != 2 or sys.argv[1] in ("-h", "--help"):
display_usage()
sys.exit(1)
directory_path = sys.argv[1]
delete_empty_directories(directory_path)
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment